'Pseudodiagnosticity' in an idealized medical problem-solving environment

Area of Science:

• Cognitive Psychology
• Medical Education
• Clinical Decision-Making

Background:

• Accurate differential diagnosis is crucial for effective patient care.
• Bayes' theorem provides a normative framework for updating diagnostic probabilities with new evidence.
• Cognitive biases can impede rational decision-making in clinical settings.

Purpose of the Study:

• To investigate how senior medical students select symptom information when evaluating two plausible diagnoses.
• To identify the prevalence of the pseudodiagnosticity effect in this student population.
• To explore the cognitive underpinnings of biased information seeking in diagnostic reasoning.

Main Methods:

• Sixty-five senior medical students were presented with hypothetical patient cases.
• Participants chose symptom information to differentiate between two potential diagnoses.
• Data selection patterns were analyzed for adherence to Bayesian principles versus biased information seeking.

Main Results:

• 83% of students failed to select information required for Bayesian computation.
• A strong tendency to seek data relevant to a single disease was observed.
• Information relevant to alternative diagnoses was frequently ignored, demonstrating pseudodiagnosticity.

Conclusions:

• Senior medical students exhibit a significant pseudodiagnosticity effect, hindering optimal differential diagnosis.
• The difficulty in simultaneously evaluating single symptoms against single diagnoses contributes to this bias.
• Medical educators should incorporate demonstrations of pseudodiagnosticity to improve students' diagnostic reasoning skills.